INDEX

Explanations

specific abbreviations related to law enforcement agencies and locations

oai_token-act-pair · gpt-3.5-turbo

abbreviations and acronyms related to organizations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 è£ıè¦ļéĨĴ

-0.77

andise

-0.76

ãĤ¹ãĥĪ

-0.70

 odds

-0.68

immer

-0.68

ãĤ¤ãĥĪ

-0.67

iously

-0.67

oulos

-0.66

 guiActive

-0.64

hetti

-0.63

POSITIVE LOGITS

VICE

1.10

FAULT

1.06

DE

1.04

leted

0.96

POS

0.93

FINE

0.91

ATH

0.86

FIN

0.84

forestation

0.83

VE

0.82

Activations Density 0.005%

specific abbreviations related to law enforcement agencies and locations

abbreviations and acronyms related to organizations

No Comments

No Known Activations